NoteBot.AI

An AI-powered study assistant that ingests documents, generates flashcards, answers questions, and helps you study smarter — all from a clean web interface.

---

Features

• Document Upload — PDF, DOCX, TXT, images (OCR), audio/video (Gemini transcription), URLs
• AI Chat — Ask questions about your uploaded documents, get cited answers
• Flashcards — Auto-generated, flippable, exportable as JSON or CSV
• Study Tools — Study guide, briefing doc, FAQ, timeline, practice questions with answer checking
• Important Points — Extracted key points with Wikipedia verification
• Calendar — Weekly planner with conflict detection and weekend summaries
• Web Mode — Summarise any URL and browse your web history

---

Tech Stack

Layer	Technology
Backend	FastAPI + SQLAlchemy (SQLite)
AI Engine	Google Gemini 2.5 Flash
Parser	pypdf, python-docx, Pillow, pytesseract, Gemini vision
Frontend	Vanilla HTML/CSS/JS (dark theme SPA)
Testing	pytest + Hypothesis (property-based)

---

Project Structure

NoteBot-AI/
├── app/
│   ├── db/
│   │   ├── database.py       # SQLAlchemy engine + session
│   │   └── init_db.py        # Table creation on startup
│   ├── models/
│   │   └── models.py         # ORM models (7 tables)
│   ├── routers/
│   │   ├── upload.py         # POST /api/upload/file|url
│   │   ├── chat.py           # POST /api/chat/query
│   │   ├── flashcards.py     # GET/POST /api/flashcards/
│   │   ├── important_points.py
│   │   ├── study_tools.py    # study guide, FAQ, timeline, Q&A
│   │   ├── calendar.py       # weekly planner
│   │   └── web_mode.py       # URL summariser
│   ├── services/
│   │   ├── parser.py         # multi-format document parser
│   │   └── ai_engine.py      # Gemini AI wrapper
│   └── main.py               # FastAPI app + error handlers
├── frontend/
│   ├── index.html
│   ├── style.css
│   └── app.js
├── tests/
│   ├── test_db_properties.py
│   └── test_parser_properties.py
├── .env
├── .gitignore
└── requirements.txt

---

Setup

1. Clone and create virtual environment

git clone <repo-url>
cd NoteBot-AI
python -m venv venv
venv\Scripts\activate        # Windows
# source venv/bin/activate   # Mac/Linux

2. Install dependencies

pip install fastapi uvicorn sqlalchemy pypdf python-docx Pillow pytesseract \
            hypothesis pytest python-multipart pydantic google-genai python-dotenv

3. Get a free Gemini API key

1. Go to aistudio.google.com/apikey
2. Sign in with any Google account
3. Click Create API key in new project
4. Copy the key

4. Configure environment

Create a .env file in the project root:

GEMINI_API_KEY=your_key_here

5. Run the server

venv\Scripts\uvicorn.exe app.main:app --reload

Open http://localhost:8000 in your browser.

---

API Reference

The full interactive API docs are available at http://localhost:8000/docs.

Method	Endpoint	Description
POST	/api/upload/file	Upload a file
POST	/api/upload/url	Ingest a URL or YouTube link
GET	/api/upload/documents	List all documents
DELETE	/api/upload/documents/{id}	Delete a document
POST	/api/chat/query	Ask a question
GET	/api/chat/history	Chat history
GET	/api/flashcards/{doc_id}	Get/generate flashcards
POST	/api/flashcards/{doc_id}/regenerate	Regenerate flashcards
GET	/api/flashcards/{doc_id}/export	Export JSON or CSV
POST	/api/study/study-guide	Generate study guide
POST	/api/study/faq	Generate FAQ
POST	/api/study/timeline	Extract timeline
POST	/api/study/questions	Generate practice questions
POST	/api/study/answer-check	Evaluate an answer
GET	/api/calendar/week	Get week entries
POST	/api/calendar/entry	Add calendar entry
POST	/api/web/summarise	Summarise a URL
GET	/api/web/history	Web browsing history
DELETE	/api/web/history	Clear web history

---

Running Tests

venv\Scripts\pytest.exe tests/ -v

3 property-based tests using Hypothesis:

• Document metadata round trip (100 examples)
• Parser TXT round trip (25 examples)
• Unsupported format error (25 examples)

---

Supported File Formats

Format	Method
.txt	stdlib
.pdf	pypdf
.docx	python-docx
.png .jpg .jpeg .webp .bmp .tiff	pytesseract OCR + Gemini vision fallback
.mp3 .wav .mp4	Gemini Files API transcription
URLs	httpx + BeautifulSoup
YouTube	youtube-transcript-api

Note: pytesseract requires the Tesseract binary installed on your system.

---

Notes

• The Gemini free tier allows ~20 requests/day on gemini-2.5-flash. Quota resets daily at midnight Pacific time.
• Never commit your .env file — it's in .gitignore.
• Switch DATABASE_URL env var to a PostgreSQL connection string for production.